Exhaustive Estimation of the Chemical Properties of Andosol in Nagano Prefecture Using Low-frequency Electromagnetic Sensor

Abstract

The upland soil cation exchange capacity (CEC); the amounts of Fe, Al, and Si; the ratio of Al/Si; the amounts of Ca, K, Na, Mg, P; available phosphoric acid (Av-P); total carbon (C), total nitrogen (N); and available nitrogen (Av-N) in upland soil were estimated by analyzing the spectra emitted by the applied low-frequency electromagnetic wave with an alternating current electromagnetic (ACE) sensor. Thirty samples of Andosol in Nagano prefecture were used in our chemo-physical experiments. Real and imaginary data of low-frequency electromagnetic waves were obtained at intervals of 5kHz in the frequency range from 10kHz to 200kHz using the ACE sensor. A leave-one-out cross-validation procedure based on the whole data set were used to evaluate the predictive ability of the Partial Least Squares regression models. The accuracy of the calibration models were evaluated using the coefficient of determination and root mean square error values. The accuracy of estimation for the CEC, Fe, Al, Si, Ca, K, Na, C, N, P, Av-P Av-N and SI/Al ratio was high. Low-frequency spectroscopic analysis using the ACE sensor proved capable of quickly estimating these chemical traits, and suggested to be useful in providing information regarding the nutritional condition of plants and improving the soil fertility diagnosis.